Orthorectification of KOMPSAT Optical Images Using Various Ground Reference Data and Accuracy Assessment

Lee, Kwangjae,Kim, Eunseon,Kim, Younsoo Hindawi Limited 2017 Journal of sensors Vol.2017 No.-

<P>Along with the appearance of high resolution satellite images, image correction using Rational Polynomial Coefficients (RPCs) has become common. Location accuracy of Korea Multipurpose Satellite (KOMPSAT) standard images is still not adequate, so, in order to leverage the KOMPSAT images for applications such as mapping and change detection, it is necessary to orthorectify the images. In this study, using updated RPCs, we performed orthorectification of KOMPSAT-2, KOMPSAT-3, and KOMPSAT-3A images using various data. Through this study, we discovered that the orthorectification result using precise Ground Control Points (GCPs) and Digital Elevation Model (DEM) is the best, but it was found that the correction results through image matching are also excellent. In particular, it was confirmed that orthoimages with a planimetric accuracy around 3 m (Root Mean Square Error (RMSE)) can be generated by using well-known matching algorithms with open data such as OpenStreetMap (OSM) and Shuttle Radar Topography Mission (SRTM) DEM, which can be acquired by anyone. Although the accuracy was low in some mountainous terrain, it was confirmed that it could be used for generating KOMPSAT orthoimages using open data. This paper describes the results for orthorectifying high resolution KOMPSAT optical images using various reference data.</P>

Measurement of thermal conductivity for single and bi layer materials using the photothermal deflection method

Kwangjae Lee,Hyunjung Kim,Byeongnam Jo,Jaisuk Yoo,Junone Kang,Pilsoo Jeon 한국물리학회 2007 THE JOURNAL OF THE KOREAN PHYSICAL SOCIETY Vol.51 No.I

In this study, the photothermal deflection method is developed to obtain the thermal conductivity of single- and bilayer materials as a non-contact method. The theoretical model incorporates a single layer and bilayer model to make the developed method capable of being applied to various materials. The photothermal deflection technique measures the deflection of the probe beam which is caused by the index-of-refraction gradient of the medium above the surface of the specimen when the energy of the modulated pump beam is absorbed by the specimen and then a temperature gradient is formed in the surroundings of the specimen. Then, the quantitative thermal conductivity is determined by comparing the measured phase difference with the theoretical value. The measured specimens are pure and bulk metals with purity over 99.98 \%, bilayer materials such as silicon wafers and thin-film metal deposited in fused silica, and oxide thin film deposited in silicon wafer. The differences of thermal conductivity between thin film and bulk for the same material are within 2 \% when the thin- film materials are optically thick. The same applies to the thermal conductivities of optically thin materials such as p- and n-type silicon wafer.

A remote pointing device by using modulated IR signals with a 2-D striped pattern

Kwangjae Lee,Yong Sin Kim,Yong Wang,Suki Kim,Kwang-Hyun Baek IEEE 2013 IEEE TRANSACTIONS ON CONSUMER ELECTRONICS - Vol.59 No.3

<P>Handheld pointing devices are increasingly in demand in digital TV market to provide intuitive human-device interfaces (HDIs). HDIs with directional buttons, joysticks, and trackballs are not very intuitive owing to their limited motions. To overcome this problem, cameras, gyroscopes, or ultrasonic sensors have been incorporated into HDIs at the expense of higher power consumption. In this paper, a low-power intuitive pointing device is presented. The transmitter of the proposed device consists of a microcontroller, two infrared light-emitting diodes, and optical components. The receiver comprises a photodiode module and a motion estimation block. The proposed device reduces power consumption by 71.3% and 23.6% as compared with the conventional camera- and gyroscope-based HDIs, respectively. Test results and performance evaluations are presented for validation.</P>

Stochastic Economic Model Predictive Control for Energy Management of a Grid-Connected DC Microgrid with Gaussian Process Time Series Forecasting

Kwangjae Lee,Jinsung Kim,Kwang-Ki K. Kim 제어로봇시스템학회 2020 제어로봇시스템학회 국제학술대회 논문집 Vol.2020 No.10

중국(中國) 상장기업(上場企業) 회계이익(會計利益)의 질(質) 실증분석(實證分析)

李光宰 ( Lee Kwangjae ) 고려대학교 중국학연구소 2020 中國學論叢 Vol.0 No.67

I have empirically examined the difference of earnings(NI_t) and cash flow (CFO_t) persistence, along with the difference in explanatory powers of NI_t and CFO_t on the market value(MV_t) and intrinsic value(IV_t) between the firms of mainland China listed in Shanghai and Shenzhen, and those listed in Hong Kong and Korea stock exchanges, using ex-post intrinsic value model suggested by Subramanyam and Venkatachalam(2007). My findings are as follows. Firstly, the results of both Ni_t+n and CFO_t+n regressions on NI_t and CFO_t respectively show that adjusted R²s and regression coefficients(bi) of the firms listed in Shanghai and Shenzhen stock exchanges are notably lower in each subsequent period, and decrease substantially faster for the next three periods(n+1~n+3) than those listed in Hong Kong and Korea stock exchanges. Secondly, the results from MV_t regressions on NI_t and CFO_t show that adjusted R²s of the listed firms of Shanghai and Shenzhen are remarkably lower than those of Hong Kong and Korea. Thirdly, the IV_t regressions on NI_t and CFO_t have also resulted significantly lower adjusted R s for the firms of Shanghai and Shenzhen than those of Hong Kong and Korea. These findings are very well compatible with my predictions that both Shanghai and Shenzhen firms’ persistence of earnings(NI_t) and cash flowCFO_t), and their explanatory powers for not only future earnings(NI_t+n) and cash flow(CFO_t+n) but also market value(MV_t) and ex-post intrinsic value(IV_t) are significantly lower than those of Hong Kong and Korea firms. My findings are no little important, providing an empirical evidence on which the Chinese government may set new accounting rules with higher earnings quality standards.

초과리익모형(超過利益模型)을 이용(利用)한 중국기업(中國企業)의 시장배수(市場倍數) 분석(分析)

李光宰 ( Lee Kwangjae ) 고려대학교 중국학연구소 2019 中國學論叢 Vol.0 No.64

This paper has empirically evaluated the chinese stock bubble during 2010∼ 2015, measuring the fundamental value (V) and PVR (price to value ratio) of listed firms on three chinese stock exchanges in Hong Kong, Shanghai and Shenzhen with residual income valuation model suggested by Feltham and Ohlson (1995). Along with the analysis of popular market multiples in investment practice such as PBR (price to book value ratio) and PER (price to earnings ratio), the difference of PVR distributions among those three stockexchanges will tell us whether the stock bubble exists in mainland China, and how serious it is now. The empirical findings are as below. Firstly, the average V of Hong Kong is 52.9% and 632.4% bigger than those of Shanghai and Shenzhen, whereas their PVR’s are much higher than Hong Kong’s by 16.8 and 15.1 times, respectively. This implies that very serious stock bubbles exist in mainland China, and the chinese stock on average is over-valued by 15∼16 times than Hong Kong’s. Secondly, the average PBR and PER of Shanghai are 4.86 and 4.477 times higher, and Shenzhen’s are also 4.55 and 4.484 times higher than Hong Kong’s, which still confirms the existence of serious stock bubble in mainland China. Finally, between two stock exchanges in mainland, Shenzhen stocks have been more bubbled than Shanghai stocks, especially in recent years of my test, during 2013∼2015, with an apparent increasing trend. My findings support the warnings from global consulting firms including Mckinsey that the stock and real estate bubbles in China could trigger a disastrous world financial crisis, and they may not be just over-spoken.

중국 상장기업의 보수주의회계성향 분석

이광재(Lee, Kwangjae) 한국국제회계학회 2017 국제회계연구 Vol.0 No.72

본 논문은 기업의 장단기 보주주의성향지표인 Ci와 Ct 지수를 사용하여, 홍콩, 상해, 심천 등 중국 3대 증권거래소 상장기업의 보수주의회계성향을 측정하고, 이들 지수에 대한 2개 회계위험척도(accounting risk measure)를 포함한 5개 위험지표와 4개 성장성지표(growth factor)의 회귀분석을 통해, 각 증권거래소별 단기적 회계위험(Ct)과 장기적 이익성장성(Ci)의 차이를 분석하였다. 본 논문의 분석결과를 요약하면 다음과 같다. 첫째, Ct, Ci 지수 모두 거래소별로 유의한 차이를 보였다. 특히 총발생액의 기간차이 Ct(T)는 홍콩, 상해, 심천, 비영업발생액의 기간차이 Ct(NOP)는 심천, 홍콩, 상해, Ci지수는 모두 심천, 상해, 홍콩 순으로 컸다. 둘째, accounting beta를 의미하는 ABETA는 상해, 심천, 홍콩, 분기회계이익의 표준편차로 측정한 ASD는 상해, 홍콩, 심천의 순으로 컸으나, 그 차이는 대부분 유의하지 않았다. 셋째, 2개 Ct지수와 5개 위험변수의 상관성 분석결과, Ct(T)와 BTM(book-to-market ratio), ABETA, Ct(NOP)와 BETA(market beta)가 각각 유의한 양의 상관성을 보였다. 넷째, 2개 Ci지수와 4개 성장성변수의 상관성 분석결과, 홍콩과 상해는 Ci지수와 총자산변동률을 의미하는 ΔTA만 유의한 양의 상관성을 보인 반면, 심천은 4개 성장성변수 모두가 Ci지수와 유의한 양의 상관성을 보였다. 다섯째, 2개 Ct지수에 대한 5개 위험변수의 회귀분석결과, 홍콩의 BTM, 심천의 BETA와 ABETA가 유의한 양의 계수를 보였다. 마지막으로, 2개 Ci지수에 대한 4개 성장성변수의 회귀분석결과, 특히, ΔTA와 RE(이익잉여금유보율)가 유의한 양의 회귀계수를 보였다. 이러한 분석결과는, Ct에 반영된 보수주의회계의 단기적 이익변동효과는 홍콩거래소가 상대적으로 큰 반면, Ci에 반영된 장기적 이익성장성은 심천거래소가 가장 높다는 사실과 함께, 본 논문에서 선정한 위험과 성장성 변수들이 각각 Ct와 Ci 지수가 포착하는 기업의 단기적 회계이익 변동위험과 장기적 이익성장성을 효과적으로 설명한다는 사실을 보여준다. This paper has empirically examined the difference of short-term earnings variation and long-term earnings growth among three major stock exchanges (Hong Kong, Shanghai, Shenzhen) in China, through the regressions of Ct and Ci scores, respectively, on 5 risk factors including 2 accounting risk measures (accounting beta and standard deviation of quarterly earnings), and 4 earnings growth factors. The Cscores I adopted had developed by Lee (2008), which proved Ct and Ci respectively to be an effective proxy of short-run earnings variation and long-run earnings growth resulted from conservative accounting. My empirical results are as follows. Firstly, Ct(T) scores representing firm-differences in total accruals are ranked in order of Hong Kong, Shanghai, Shenzhen, whereas Ct(NOP) capturing firmdifferences in non-operating accruals are ranked in order of Shenzhen, Hong Kong, Shanghai, which means that Hong Kong Exchange records the highest short-term variability of accounting earnings, and Shenzhen Exchange shows the highest earnings growth among chinese stock exchanges. Secondly, ABETA and ASD respectively representing accounting beta and standard deviation of earnings are ranked in order of Shanghai, Shenzhen, Hong Kong, and Shanghai, Hong Kong, Shenzhen, respectively. Thirdly, Ct(T) and Ct(NOP) respectively shows strong positive correlations with BTM (book-to-market ratio), BETA (market beta), and ABETA. Fourthly, Shenzhen Exchange reports significantly positive correlations with all four growth factors adopted in this paper, while Hong Kong and Shanghai Exchanges report the positive correlation only with ΔTA (annual changes in total assets). Finally, regressions of Ct and Ci respectively on 5 risk and 4 growth factors show significantly positive coefficients of BTM in Hong Kong, BETA in Shenzhen, and ΔTA, RE (retained earnings to total assets ratio) in all three exchanges. These results imply that there are significant C-score differences among three chinese stock exchanges, and the risk and growth factors the paper adopted have effectively explained the listed firms’ variation in Ct and Ci scores, among three stock exchanges, reflecting their short-run earnings risk and long-run earnings growth, respectively.

`교육정보미디어연구`에 나타난 온라인 협력학습에서의 학습자간 상호작용분석 방법 동향

이광재 ( Kwangjae Lee ),이영주 ( Youngju Lee ) 한국교육정보미디어학회(구 한국교육정보방송학회) 2017 교육정보미디어연구 Vol.23 No.1

본 연구에서는 온라인 협력학습환경에서 학습자간 상호작용 분석을 실행한 연구 사례들을 수집하여 상호작용 분석 유형을 분류하고, 분석 방법의 동향 및 흐름을 파악하고자 한다. 본 연구는 1996년에서 2015년까지 20년간 발간된 “교육정보미디어연구” 학회지 1996년 2권 1호부터 2015년 21권 4호까지 게재된 논문을 검토한 결과 총 35편의 논문을 대상으로 선정하였다. 연구 사례들에 사용된 학습자간 상호작용 분석방법의 유형을 6가지 범주(참여도 분석, 내용분석, 사회연결망분석, 검사도구 활용, 과정의 시각화, 혼합방법)로 유목화하여 분류하였다. 연구 결과, 2010년도 이전까지는 학습자 참여도 분석방법과 내용분석방법이 학습자간 상호작용분석에서 주된 연구방법으로 사용되었으나 2011년 이후로는 사회연결망분석방법, 혼합분석 방법 등 6가지 유형별로 고르게 활용되는 분포를 보였다. 분석대상논문의 57%는 내용분석을 37%는 참여도 분석을 활용하였다. 내용분석방법은 학습자들의 메시지를 통해서 학습이 이루어진 과정을 알 수 있기 때문에 상호작용의 깊이 있는 질적인 분석이 가능한 반면에 메시지 간의 관계가 의미하는 것을 파악하기 힘들다. 학습자 참여도 분석은 객관적 측정이 가능하나 상호작용의 과정적 흐름 및 관계 등 질적인 측면을 파악하기 어렵다. 사회연결망 분석은 11%를 차지하였는데, 학습자간의 관계와 구조를 파악할 수 있는 장점이 있지만 상호작용의 내용과 질에 대한 면밀한 분석이 어렵다는 단점이 있다. 본 연구의 결과는 이러닝 뿐만 아니라 SNS기반을 포함한 모바일 학습 환경에서 급증하는 온라인 협력학습 환경에서 학습자간 상호작용 분석에 관심이 있는 많은 연구자와 교수자에게 유용한 정보를 제공하는데 의의가 있다. In this study, we investigated patterns and trends on analysis of learners` interactions in the online collaborative learning environments by examining previous studies using analysis of learners` interactions. We gathered a total number of 35 studies which applied analysis of learners` interactions and published in “The journal of Educational Information and Media” from Issue 2, Volume 1 of 1988 to Issue 4, Volume 21 of 2015. We analyzed the studies and sorted them into 6 categories(participation rate analysis, content analysis, social network analysis, survey use, visualization of interaction process, mixed methods). The results indicated that participation rate analysis and content analysis was primarily used until 2010. Since 2011, six analysis methods were distributed evenly. 57% of the studies used content analysis and 37% of the studies utilized participants rate analysis, which accounted for the great portion of the target studies. Content analysis enables us to conduct deep, qualitative analysis with examination of the learners` messages which reflect learning process. The participation rate analysis enables us to perform objective measure but it is difficult to examine qualitative aspects such as process, relationship of interactions. The social network analysis occupied 11% of the studies and it enables researchers to examine learners` relationships and structure of interactions. The results of the study provide researcher and instructors interested in learners` interactions analysis in online collaborative learning including mobile based instruction with useful information.

다중 센서들을 이용한 다양한 화재 상황의 위험도 추정 시스템 개발

이광재 ( Kwangjae Lee ),이연성 ( Youn-sung Lee ) 한국센서학회 2016 센서학회지 Vol.25 No.6

In this paper, a fire detection system based on quantitative risk estimation is presented. Multiple sensors are used to build a comprehensive indicator that represents the risk of fire quantitatively. The proposed fire risk estimation method consists of two stages which determines the occurrence of fire and estimates the toxicity of the surveillance area. In the first stage, fire is reliably detected under diverse fire scenarios. The risk of fire is estimated in the second stage. Applying Purser`s Frac-tional Effective Dose (FED) model which quantitates harmfulness of toxic gases, the risk of the surveillance area and evac-uation time are calculated. A fire experiment conducted using four different types of combustion materials for the verification of the system resulted in a maximum error rate of 12.5%. By using FED calculation and risk estimation methods, the proposed system can detect various signs of fire faster than conventional systems.

무선 메쉬 네트워크에서 효과적인 멀티캐스트를 위한 무선 자원 할당 기법

이종관(Lee Jong-Kwan),박상재(Park Sang-Jae),이재준(Lee Jae-Joon),임재성(Lim Jae-Sung),임광재(Lim Kwangjae),이안석(Lee Anseok),신우람(Shin Wooram) 한국통신학회 2013 한국통신학회 학술대회논문집 Vol.2013 No.1